8 pages

Effect of the potential on the electrochemically induced ageing of polyaniline films

Please download to get full document.

View again

of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Effect of the potential on the electrochemically induced ageing of polyaniline films
  Effect of the potential on the electrochemically induced ageing of polyaniline films Waldemar A. Marmisollé, M. Inés Florit ⇑ , Dionisio Posadas Instituto de Investigaciones Fisicoquímicas Teóricas y Aplicadas (INIFTA), Facultad de Ciencias Exactas, Universidad Nacional de La Plata, CCT La Plata-CONICET. Sucursal 4,Casilla de Correo 16, 1900, La Plata, Argentina a r t i c l e i n f o  Article history: Received 17 August 2011Received in revised form 6 January 2012Accepted 18 January 2012Available online 2 February 2012 Keywords: PolyanilineElectrochemical ageingPotential dependence a b s t r a c t The electrochemically induced ageing of polyaniline films, as a function of the ageing potential, wasinvestigated employing films of different thicknesses in 3.7 M H 2 SO 4  solutions. The ageing process wasmonitored by the change of the voltammetric response in a subsequent positive going potential scan afterthe ageing. It is observed that the ageing process, at more positive potentials, involves increasingamounts of non-reduced polymer. Also, it is observed that the rate of ageing decreases as the ageingpotential increases. The changes in the voltammetric response after ageing is interpreted through amodel for the voltammetric response, previously developed, that includes capacitive currents and inter-actions between the redox centers. The analysis shows that neither the capacitive parameters nor thetotal faradaic charge change during the ageing. On the other hand, only the interaction energy betweenthe reduced centers change during the ageing. This fact leads to a narrowing of the formal redox potentialdistribution that can be characterized by the apparent number of exchanged electrons. The decrease of the interaction energy between reduced centers is consistent with the reduction of the free energy of the polymer during the ageing. The kinetics of ageing is analyzed through an Elovich type of equationfrom which  k 0 , the  pseudo  zero order rate constant and  b , a parameter that indicates how fast the activa-tion free energy changes with the extent of the process, can be obtained.   2012 Published by Elsevier B.V. 1. Introduction The term  physical ageing   is employed to describe a process inwhich amorphous materials irreversibly go through slow struc-tural modifications after being cooled below the glass transitiontemperature [1]. During the physical ageing, the variation of sev-eral properties of the material (such as specific volume, enthalpyand entropy) is linear with the logarithm of the ageing time [2],thus suggesting that only structural changes occurs during the pro-cess. Of course, heating the material again erases all memory of theageing process. When certain electrochemically active polymers,i.e. polymers that can reversible oxidized and reduced, mostly con-ducting polymers (CPs), are submitted to a suitable reductive po-tential, undergo a process that have many similarities with thephenomenon of physical ageing. Thus, after ageing the subsequentelectrochemical oxidation shows changes in the current peak andthe peak potential, a decrease in the specific volume, changes inthe ESR and UV–Vis response, etc. The most striking similarity be-tween physical and ‘‘electrochemical’’ ageing is that the mentionedchanges are linearly related to the logarithm of the ageing time.This particular kinetic feature indicates that both phenomena(physical and electrochemical) are self-inhibiting. Here again, thepolymer oxidation erases all traces of the ageing process. Althoughthis electrochemical ageing has received many names such as‘‘slow relaxation’’, ‘‘memory effect’’, and ‘‘first cycle effect’’, we pro-posed [3] the term ‘‘Electrochemically Induced Ageing’’ ( EIA ) to re-fer to this phenomenon in order to enforce the idea that thecondition at which the polymer ages is achieved by an electro-chemical perturbation and that is a physical ageing process.Manyworkers have studiedthe  EIA  [3–45]. Severalexplanationsand models have been proposed to explain this behavior [24–34].Besides voltammetry [3–23], electrochemical ageing have beenstudied by numerous ‘‘in situ’’ techniques: ERS [9], injection/ejec-tion of protons [35], volume changes [36,37], UV–Vis spectra [38–40], XPS [41], EIS [42] and other studies such as the effect of  the pH [43], the temperature[44], and the effect of substituents on the ageing rate [45].As it was said above, the  EIA  of CPs has been often studied mea-suring the changes in the voltammetric peak parameters  j  p  and  E   p during the first voltammetric potential scan in the positive direc-tion (the  first cycle ) after the ageing. This electrochemical responsegives a typical kinetic behavior in which the property being mon-itored (i.e.  j  p  or  E   p ) changes linearly with the logarithm of the age-ing time. This behavior indicates that the modifications of the peakparameters are related to the changes that happen during the  EIA .However, the relationship between the changes in the voltam-metric response and the ageing process is complex because theelectrochemical response of CPs contains an indeterminateamount 1572-6657/$ - see front matter    2012 Published by Elsevier B.V.doi:10.1016/j.jelechem.2012.01.017 ⇑ Corresponding author. E-mail address: (M.I. Florit). Journal of Electroanalytical Chemistry 669 (2012) 42–49 Contents lists available at SciVerse ScienceDirect  Journal of Electroanalytical Chemistry journal homepage:  of capacitive contribution, and even the faradaic contribution can-not be considered as an ideal Nernstian process [46,47]. Previously,it was published a model that takes into account not only thecapacitive contribution, but also the presence of interactions be-tween redox centers to formally represent the voltammetric re-sponse [3,47]. In the present work this model is employed to interpret the changes in the voltammetric parameters during theageing process, focussing on the effect of the ageing potential onthe  EIA . 2. Experimental A conventional three-electrode glass cell was employed for theelectropolimerization procedure and the electrochemical measure-ments. The working electrode was a 0.05 cm diameter Au wire of 0.154 cm 2 . The auxiliary electrode was a cylindrical Pt foil. The ref-erence electrode was a Saturated Calomel Electrode (SCE). Allpotentials in the text,  E  , are referred to this electrode. Conventionalvoltammetry was performed using a Teq_02 potentiostat.The polymer films were synthesized by electropolymerizationfrom 0.5 M aniline (Fluka-Garantie, puriss. p.a. previously distilledunder reduced pressure) solutions in 3.7 M H 2 SO 4  (Backer, p.a.).The deposition procedure was performed by cycling the poten-tial at  v  = 0.100 V s  1 , between   0.200 V vs. SCE, and the potentialcorresponding to that of the beginning of the oxidation of themonomer, around 0.800 V vs. SCE, as described before [16]. Thefilm characterization was performed by voltammetry in 3.7 MH 2 SO 4  solutions by cycling the potential at  v   = 0.100 V s  1 between  0.200 V and 0.450 V.In a typical ageing experiment, the potential was cycled until asteady voltammogramwas reached. Then, during the cathodic scanthe sweep was stopped and held at the ageing potential ( E  a ) duringa variable time (0 s <  t  a  < 1000 s), being  t  a  the ageing time. Aftereach ageing time, the scan was restarted in the positive direction,recording the  first cycle .  E  a  was comprised between   0.200 V and0.000 V.The voltammetric charge of the films employed in this work, Q  T  (0.45), was determined from the integration of the anodic  j / E  profiles of the voltammetric response in the potential range com-prised between  0.200 and 0.450 V vs. SCE at  v   = 0.1 V s  1 . The vol-tammetric charges employed in this work are in the rangecomprised between 20.0 mC cm  2 <  Q  T  (0.45) < 95.0 mC cm  2 . Thevoltammetric charge will be employed as a measure of the filmthickness. This can be estimated from data previously obtainedfor Pani and other substituted arylamine polymers, by ellipsometry[48–50]. Employing the relation thickness/charge reported by Zer-bino et al., for a related polymer such as poly methylaniline [50],these film thicknesses would correspond to 1200–5700 nm. 3. Results Fig. 1 shows the steady voltammetric anodic response of a Panifilm for different negative potential limits. It is seen that the peakcurrent decreases as this limit increases. However, all the voltam-metric traces end at the same current value. This means that at theupper potential limit, the amount of oxidized polymer, which isproportional to  Q  0 (see below), is the same irrespective of the neg-ative potential limit. In turn, this means that as the negative poten-tial limit increases the amount of reduced polymer, during thecathodic scan, is slightly smaller and therefore the oxidationcharge becomes also smaller.Fig. 2 shows the voltammetric response of the first cycle afterwaiting for increasing ageing times,  t  a , at three different ageingpotentials,  E  a , for a Pani film  Q  T  (0.45) = 92.9 mC cm  2 , in 3.7 MH 2 SO 4  solution. As it can be observed, the qualitative behavior isthe same at the three potentials: the anodic peak current increaseswith the ageing time and the corresponding peak potential movesin the positive direction. Note that it does not no matter what the E  a  value is, the current at the positive potential limit is the same forthe different ageing times; what it means that the amount of oxi-dized polymer at this potential does not change after ageing. Fig. 3shows the evolution of those peak parameters as a function of thelogarithm of the ageing time for a Pani film aged at differentpotentials. EIA  is characterized by showing a nearly linear dependence of   j  p and  E   p  on the logarithm of the ageing time,  t  a  [47 and referencestherein]. As it was mentioned at the introduction, this kineticbehavior is also characteristic of the physical ageing of amorphoussolids. In the literature of physical ageing, the rate of ageing is de-fined as the slope of these semi-logarithmic plots [1,2]. So, it is pos-sible to define a rate of   EIA  in terms of the anodic peak current inthe same way as it is defined for the specific volume or fluores-cence emission for the physical ageing [1,2,51–53]. r   ja  ¼  1  j  p ; 0 @   j  p @  ln ð t  a Þ ð 1 Þ where  j  p ,0  is the peak current for the steady voltammogram.Although it is not a true rate in a kinetic sense, it results to be a use-ful parameter that allows characterizing and comparing the tempo-ral evolution of the  EIA . In Fig. 4 are shown the values of this rate,for different films at different ageing potentials. The ageing rate de-fined by Eq. (1) decreases as the ageing potential is increased. Tointerpret the changes observed in the voltammetric response, inthe frame of the physicochemical processes involved, a suitablemodel is needed. The analysis of these experimental data withinthe model previously presented is the subject of the next section[47]. 4. Discussion 4.1. A Model for the voltammetric response and the EIA The voltammetric response of CPs is complex because the totalcurrent contains a capacitive current contribution that cannot beseparatedfrom the faradaic currentjust employingthe voltammet-ric information.In the model previously presented [47], the voltammetric totalcharge is considered to be composed by a capacitive charge as wellas a faradaic charge. The capacitance is considered to be propor- E / V -      j    /  m   A  c  m   -   2 05101520 Ea= -0.20VEa= -0.18VEa= -0.16VEa= -0.14VEa= -0.12VEa= -0.10VEa= -0.08VEa= -0.06VEa= -0.04VEa= -0.0.2VEa= -0.00V Fig. 1.  Stationary anodic  j / E   response of a Pani film for different negative potentiallimits ( E  a ) (see the box) in 3.7 M H 2 SO 4 .  Q  T  (0.45) = 21.5 mC cm  2 ,  v   = 0.100 V s  1 . W.A. Marmisollé et al./Journal of Electroanalytical Chemistry 669 (2012) 42–49  43  tional to the fraction of oxidized material and independent on theapplied potential [46].Under these conditions, the voltammetric current obtained for alinear potential scan results to be [46,47].  j T  ð E  Þ v  C  d ¼ ð E   þ r Þ  j  f  ð E  Þ v  Q  0  þ  Q   f  ð E  Þ Q  0  ð 2 Þ where  v   is the scan rate,  j  f   and  Q   f   are the faradaic current and chargerespectively; and  Q  0 is the faradaic charge involved in the completeoxidation of the film. Then, the ratio  Q   f  / Q  0 =  x  is the fraction the re-dox centers in the oxidized state. Finally,  C  d  and  r  are parametersrelated to the capacitive response [46]. The first one is thecapacitance when the whole polymer film is oxidized. That beingso,  C  d  is proportional to the amount of the redox centers in the film, C  d  =  aQ  0 , and the height of the capacitive plateau after the voltam-metric peak is  vC  d . As it was mentioned in reference to Figs. 1 and2, it means that if the current at the plateau remains constant, inde-pendent of   E  a  and  t  a , also  C  d  has the same value. And, as it is veryunlikely that the proportionality constant,  a , and the charge,  Q  0 , ex-actly compensates to each other, it must be concluded that both of them,  Q  0 and  a , remain constant after ageing. This reasoning alsoapplies to the experiments in which the cathodic potential limit ischanged (see Fig. 1). On the other side, within this formalism r  =  a  1  E   Z  ; where  E   Z   is the extrapolated potential at which thecharge capacitance of the polymer film should be zero. Although a complete separation of faradaic and capacitive con-tributions is not possible from the voltammetric response, bothcapacitive parameters,  C  d  and  r , are experimentally accessibleand they can be evaluated from the knowledge of the integratedvoltammetric charge as a function of potential [3,47].Figs. 5 and 6 show the temporal variation of these two capaci-tive parameters during the ageing of three different Pani films atdifferent ageingpotentials.As  C  d  is proportional to  Q  0 , it is also pro-portional to the voltammetric charge  Q  T  . However,  C  d  does not sen-sibly change either with the potential at which the scan is startedor with the ageing time. As it can be seen,  r  slightly diminishes asthe film thickness increases and during the ageing, but it does notdepend much on the ageing potential.Although the capacitive contribution does affect the height andposition of the voltammetric peak, its effect is not very important.Moreover, as the capacitive parameters change very little during E   / V E   / V E   / V -0.2-    j   /  m   A  c  m   -   2 -40-20020406080100- Fig. 2.  Current–potential response after different ageing times, t a /s = 0, 6, 12, 24, 48, 96, 192, 384, 768 at  E  a  =   0.200 V (left);   0.100 V (center); and 0.000 V (right). Q  T  (0.45) = 92.9 mC cm  2 . Arrows indicate increasing ageing times. ln ( t  a / s)      j     p    /  m   A  c  m   -   2 141618202224262830 (a) ln ( t  a  / s) 12345671234567      E     p    /   V 0.2250.2300.2350.2400.2450.2500.255 (b) Fig.3.  Anodicpeak current density (a) and anodic peakpotential (b) as afunction of the logarithm of the ageing time for different ageing potentials,  E  a / V  , ( d )  0.200 V;(O)   0.160; ( . )   0.120; ( D )   0.080; ( j )  0.040; ( h ) 0.000.  Q  T  (0.45) = 21.5 mCcm  2 . Fig. 4.  Ageing rate in terms of the peak current, as a function of the ageing potentialfor different Pani films.  Q  T  (0.45)/mC cm  2 : ( d ) 21.5; (O) 33.3; ( . ) 92.9.44  W.A. Marmisollé et al./Journal of Electroanalytical Chemistry 669 (2012) 42–49  the  EIA , the modifications of the voltammetric peak features mustbe due to the variations of the faradaic process during the first po-sitive potential scan after ageing.In order to analyze the faradaic contribution during the  EIA , asuitable model for the electrochemical transformation of the redoxcenters must be considered. An expression for the total voltam-metric current as a function of the fraction of the redox centers,  x , can be derived, assuming that the redox process is electrochem-ically reversible during the potential scan and taking into accountthe existence of interactions between the redox centers within the mean field approximation  [47].  j T  v  C  d ¼  nF RT  ð E   þ r Þð 1   x Þ  x ½ 1  ð n Ox  þ  n R Þð 1   x Þ  x  þ  x  ð 3 Þ In Eq. (3) the dependence of the potential on  x  is implicit be-cause from the model the dependence of   E   on  x  results: E   ¼  E  o 0   RT nF   ½ð n Ox  þ  n R Þ  x    n R    RT nF   ln 1   x x  ;  ð 4 Þ where  E  0 0 is the formal redox potential and  n Ox  and  n R  are dimen-sionless interaction parameters.  n Ox  is proportional to the differenceof mean interaction energy between the pairs of neighbor redoxcenters,  Ox–R  ( e OR ) and  Ox – Ox  ( e OO ) and, correspondingly,  n R  is pro-portional to the difference in the mean interaction energy between Ox – R  and  R – R  pairs ( e RR ) [3,47] n R  ¼  cN   A v  ð e RR   e OR Þ RT   ð 5 Þ and n Ox  ¼  cN   A v  ð e OO   e OR Þ RT   ð 6 Þ where  c   is the mean number of redox neighbor centers. The peak potential is found from the condition that the deriva-tive of the right-hand side of Eq. (3) with respect to the potential isequal to zero, and then Eq. (3) gives the peak current [47]. Then, the expression for the oxidized fraction at the peak potential,  x  p ,results:  x  p  ¼  12 þ  RT nF  ½ 1  ð n Ox  þ  n R Þð 1   x  p Þ  x  p  2 ð E   p  þ r Þ ð 7 Þ Here the subscript ‘‘  p ’’ indicates peak values. Having previouslydetermined  r  and  C  d , the interaction parameters can be evaluatedfrom the values of   j  p  and  E   p  by employing an iterative procedure inwhich  x  =  x  p  is considered in Eqs. (3) and (4) [47]. Following this procedure, the interaction parameters for differ-ent Pani films after different ageing times were determined. Previ-ously, it was calculated that the variations of the  n Ox  parameter arenot noticeable and the changes in the faradaic contribution can bealmost exclusively assigned to  n R  [3]. Fig. 7 shows the time varia- Ea=-0.200VEa=-0.160VEa=-0.120VEa=-0.080VEa=-0.040VEa=-0.200VEa=-0.150VEa=-0.100VEa=-0.050V ln ( t  a   / s)ln ( t  a   / s)ln ( t  a   / s) 234567234567234567      C      d    /  m   F  c  m   -   2 020406080100120140160 Ea=-0.200VEa=-0.160VEa=-0.120VEa=-0.080VEa=-0.040V Fig. 5.  The capacitive parameter  C  d  as a function of the logarithm of the ageing time, for Pani films of different thicknesses, aged at different potentials indicated in eachgraphic.  Q  T  (0.45): 21.5 mC cm  2 (left); 33.3 mC cm  2 (center); 92.9 mC cm  2 (right). Ea =-0.20VEa =-0.16VEa =-0.12VEa =-0.08VEa =-0.04VEa =-0.20VEa =-0.15VEa =-0.10VEa =-0.05V ln ( t  a  / s)ln ( t  a  / s)ln ( t  a  / s) 2345672345671234567    σ    /    V Ea =-0.20VEa =-0.18VEa =-0.12VEa =-0.08VEa =-0.04V Fig. 6.  The capacitive parameter  r  as a function of the logarithm of the ageing time, for  c   Pani films different thicknesses, aged at different potentials indicated in eachgraphic.  Q  T  (0.45): 21.5 mC cm  2 (left); 33.3 mC cm  2 (center); 92.9 mC cm  2 (right). W.A. Marmisollé et al./Journal of Electroanalytical Chemistry 669 (2012) 42–49  45  tion of this parameter for different Pani films aged at differentpotentials.It is observed that the  n R  values are smaller for the more posi-tive ageing potentials. Also, it is observed that its values increaselinearly with the logarithm of the ageing time, and that they donot depend markedly on the film thickness. Then, being  n Ox  nearlyconstant [3], and taking into account the definition of   n R , its in-crease as the  EIA  progresses means that the interaction energy be-tween reduced centers,  e RR , decreases and it is the responsible forthe change in the voltammetric response after the ageing. The de-crease of the interaction energy between reduced centers is consis-tent with the reduction of the free energy of the polymer in thereduced state during the ageing. As this form becomes more stable,its subsequent oxidation is more difficult and happens at higherapplied potentials.Eq. (4) shows that this system does not correspond to an idealNernstian one; the activity coefficients for both reduced and oxi-dized centers depend exponentially on the degree of oxidation. Inthe present model, that point is a consequence of the introductionof interactions between redox centers and it results in a formal re-dox potential distribution. In this particular case, the formal redoxpotential varies linearly with the degree of oxidation [47].The apparent number of exchanged electrons,  n ap , is a usefulparameter that describes the width of the formal redox potentialdistribution and it is defined from the slope of the plot of the deriv-ative of the potential respect to ln[  x /(1   x )]. For an ideal Nernstiansystem, this slope is   RT  / nF  , while for many systems the slope isconstant at least in a range around  x  = 0.5 but its value is differentto   RT  / nF  . In this particular case the slope of the plot  E   vs. ln[  x /(1   x )] its constant in the range 0.25 <  x  < 0.75, but is value is smal-ler than   RT  / nF   [54,55]. In general the slope may be written as: @  E  @  ln ðð 1   x Þ =  x Þ ¼   RT n ap F   ð 8 Þ By series expansion of   E   with respect to ln[  x /(1   x )] around thepoint  x  = 0.5 in Eq. (4), and equating the result to   RT  / nF  , accordingto Eq. (8), the following relation results: n ap  ¼  44  ð n R  þ  n Ox Þ n  ð 9 Þ As expected, when the interaction parameters are null, the sys-tem becomes an ideal Nernstian one and  n  =  n ap . By consideringthat in the voltammetric peak, the degree of oxidation is approxi-mately 0.5, the peak current results,  j  p  ¼  v  C  d n ap F  ð E  O þ r Þ 4 RT   þ 12 !  ð 10 Þ where  E  O =  E  o 0  ( RT  /2 nF  )( n Ox    n R ). This indicates that the peak cur-rent is approximately proportional to the apparent number of elec-trons. In Fig. 8, the time dependence of   n ap  (calculated by Eq. (9)) isshown for the same Pani films presented in the former figures. Inthe present case,  n  = 2 and  n ap  is always smaller than  n  (seeFig. 8). This means that the formal redox potential distribution iswider than in the absence of interactions. However,  n ap  increasesas the waiting time increases indicating that, as a consequence of the  EIA , the formal redox potential distribution becomes narrowerduring the ageing process. This effect is directly monitored by theincrease in the current peak. It is interesting to note that  n ap  also increases linearly with thelogarithm of the ageing time. An ageing rate in terms of   n ap  couldbe then defined in the same way it was done for the peak current, r  ð n ap Þ ¼  @  n ap @  ln  t  a ð 11 Þ The correlation between the ageing rates defined in terms of both the peak current and the apparent number of electrons isshown in Fig. 9. As it can be seen, they are well correlated in a lin-ear way indicating the peak current changes can be attributed tothe changes in the apparent number of electrons, whose temporalvariation means that the formal redox potential distribution be-comes narrower as a consequence of the  EIA .Additionally, it deserves to point out that the analysis of thecurrent peak behavior allows to determine that the major varia-tions in  n ap  are achieved at more negative values of the ageing po-tential. That is, the ageing rate increases as the potential becomesmore negative. 4.2. The kinetics of the EIA The  EIA  is characterized by a particular time behavior which isalso found in the physical ageing of amorphous solids. The extentof the process, expressed as the relative change of some particularproperty (specific volume, fluorescence emission, etc.), dependslinearly on the logarithm of the ageing time. This time dependencewas also found by Elovich when studying heterogeneous chemicalreactions. Therefore, an Elovich type of kinetics can be also used todescribe the time evolution of the  EIA  [45]. This kinetic can be rep-resented by the following rate law [56]: v   ¼  d q = dt  ¼  k 0 exp ½ð D G  0  þ  b q Þ = RT   ð 12 Þ E a =-0.20VE a =-0.16VE a =-0.12VE a =-0.08VE a =-0.04V E a =-0.20VE a =-0.15VE a =-0.10VE a =-0.05V ln ( t  a  / s)ln ( t  a  / s)ln ( t  a  / s) 234567 234567 234567     ξ    R 0123456   E a =-0.20VE a =-0.16VE a =-0.12VE a =-0.08VE a =-0.04V Fig. 7.  Time variation of the interaction parameter  n R  for three Pani films aged at different potentials indicated in each graphic.  Q  T  (0.45): 21.5 mC cm  2 (left); 33.3 mC cm  2 (center); and 92.9 mC cm  2 (right).46  W.A. Marmisollé et al./Journal of Electroanalytical Chemistry 669 (2012) 42–49
Related Documents
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!